Available online at www.sciencedirect.com

www.elsevier.com/locate/asr

ScienceDirect

Advances in Space Research 62 (2018) 2174–2186

Science exploration architecture for Phobos and Deimos: The roleof Phobos and Deimos in the future exploration of Mars

Ariel N. Deutsch a,⇑, James W. Head a, Kenneth R. Ramsley b, Carle M. Pieters a,Ross W.K. Potter a, Ashley M. Palumbo a, Michael S. Bramble a, James P. Cassanelli a,

Erica R. Jawin a, Lauren M. Jozwiak a, Hannah H. Kaplan a, Connor F. Lynch a,Alyssa C. Pascuzzo a, Le Qiao a,c, David K. Weiss a

aDepartment of Earth, Environmental and Planetary Sciences, Brown University, Providence, RI 02912, USAbSchool of Engineering, Brown University, Providence, RI 02912, USA

cPlanetary Science Institute, China University of Geosciences, Wuhan, Hubei 430074, China

Received 4 February 2017; received in revised form 4 December 2017; accepted 12 December 2017Available online 18 December 2017

Abstract

Phobos and Deimos are the only natural satellites of the terrestrial planets, other than our Moon. Despite decades of revolutionaryMars exploration and plans to send humans to the surface of Mars in the 2030’s, there are many strategic knowledge gaps regarding themoons of Mars, specifically regarding the origin and evolution of these bodies. Addressing those knowledge gaps is itself important,while it can also be seen that Phobos and Deimos are positioned to support martian surface operations as a staging point for futurehuman exploration. Here, we present a science exploration architecture that seeks to address the role of Phobos and Deimos in the futureexploration of Mars. Phobos and Deimos are potentially valuable destinations, providing a wealth of science return, as well as telecom-munications capabilities, resource utilization, radiation protection, transportation and operations infrastructure, and may have an influ-ence on the path of the martian exploration program. A human mission to the moons of Mars would maintain programmatic focus andpublic support, while serving as a catalyst for a successful human mission to the surface of Mars.� 2017 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Phobos; Deimos; Mars; Human exploration; Mission architecture

1. Introduction

As NASA plans to send humans to Mars in the 2030’s(https://www.nasa.gov/content/nasas-journey-to-mars;e.g., Head et al., 2015), it is vital to gain a further under-standing of the role of the martian moons, Phobos andDeimos, in this next stage of planetary exploration. Thesebodies may facilitate later exploration of Mars, throughexploitation of in situ resources, by providing radiation

https://doi.org/10.1016/j.asr.2017.12.017

0273-1177/� 2017 COSPAR. Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: Ariel_Deutsch@brown.edu (A.N. Deutsch).

protection, or through contributions to the continuity ofthe martian program. It is crucial that a mission architec-ture is developed to serve as a roadmap for the explorationof Mars’ moons focusing on the questions: ‘‘What is theorigin of Phobos and Deimos?” and ‘‘What is the role ofPhobos and Deimos in the future exploration of Mars?”.

Besides our Moon, Phobos and Deimos are the onlyknown natural satellites of the terrestrial planets. We seekto understand why our Moon is different than the martianmoons, and what their presence tells us about the moon-less Mercury and Venus. Understanding these satellitescan help us understand the origin and evolution of the

Fig. 1. Mars-facing side of Phobos, highlighting the linear grooves acrossthe surface of the moon and Stickney crater in the left of the image. Imageacquired by the High Resolution Stereo Camera from the Mars Expressspacecraft on 22 August 2004. Pixel resolution of �7 m. Image credit:ESA/DLR/FUB.

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2175

martian system. To date, the origins of Phobos and Deimosare still unknown. The hypothesized formation theorieshave specific and distinct implications for the possibilitiesof finding pristine martian crust, primitive material, orvolatiles (e.g., Murchie et al., 2014). Depending on theircomposition, these moons could greatly facilitate Marsexploration (e.g., Oberst et al., 2014) (Table 1).

There are many strategic knowledge gaps concerningPhobos and Deimos science. Here we design a missionarchitecture that can, first, answer the fundamental scien-tific questions of the martian moons and, second, assessto what extent Phobos and Deimos can facilitate the laterhuman exploration of Mars. This facilitation could occurin many ways (as discussed in-depth in Section 3), thereforeit is crucial to develop a framework to serve as a roadmapbefore sending missions to Phobos and Deimos.

The major initial question driving our framework has ascientific motivation: ‘‘What is the origin of Phobos andDeimos?” The second question driving our framework askshow stopping at one or both moons could complement ahuman exploration program to Mars: ‘‘What is the roleof Phobos and Deimos in the future exploration of Mars?”

To address the above questions, we outline major gapsremaining in Phobos and Deimos science that motivatefuture missions to the moons of Mars. We present a scienceexploration architecture consisting of orbiter, lander, andhuman exploration stages and discuss a frameworkdesigned to assess the role of Phobos and Deimos in thefuture exploration of Mars, specifically determining howthese bodies might aid in the Mars program.

2. Strategic knowledge gaps in Phobos and Deimos science

Many unanswered questions regarding the formationand evolution of Phobos (Fig. 1) and Deimos exist today(e.g., Basilevsky et al., 2014; Murchie et al., 2014). Herewe discuss the major gaps in the understanding of the ori-gin and evolution of Phobos and Deimos that motivate theneed for future exploration.

2.1. Origin

Many formation hypotheses have been suggested for theorigin of Phobos and Deimos. One hypothesis is that thesemoons are captured asteroid belt objects, consistent withcompositional data and optical and spectral properties ofPhobos and Deimos (e.g., Fraeman et al., 2012, 2014;Pajola et al., 2013; Pieters et al., 2014). Specifically, thelow albedos, low densities, and spectral similarities of the

Table 1Formation models of Phobos and Deimos and implications for water and car

Origin model Water content (wt%)

Co-accretion �1 (moderate) Sarafian et al. (2014)Capture 2–60+ (high–very high) Murchie et al. (2Mars impact �0.015 (low) Truong and Lee (2017)

moons to D-type asteroids are consistent with capture ofouter solar system objects that are composed of primitive,carbonaceous material (Hartmann, 1990; Burns, 1992).Additionally, Phobos and Deimos have near-identical sur-faces, so if they are captured, then they either are of thesame original composition, or they are unrelated and haveundergone extensive space weathering to produce near-identical surfaces (Murchie et al., 2014). The process ofpermanent capture, however, requires sufficient energy lossand has proven to be difficult to explain dynamically. BothBurns (1978) and Pollack et al. (1979) described qualita-tively a possible method by which to capture a single objectby aerodynamic drag in a circumplanetary envelopearound Mars. Dynamical models that explore this phe-nomenon produce a Phobos spiraling toward Mars and aretreating Deimos (Hunten, 1979; Sasaki, 1990). Whilethe physical properties of the moons suggest that theymay be dark carbonaceous asteroids that were capturedby Mars (e.g. Fraeman et al., 2012, 2014; Pajola et al.,2013; Pieters et al., 2014), the predicted orbital evolutiondue to the tides of the martian system suggests that the

bon content.

Carbon content

Low Murchie et al. (2014)014); Lee et al. (2017) High Murchie et al. (2014)

Low Murchie et al. (2014)

2176 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

moons originated on nearly circular, uninclined orbits notfar from their current positions (Burns, 1992).

Alternatively, the moons of Mars, like other regularsatellites, may be accretional products of orbiting debrisin the planetary formation process (e.g., Burns, 1986).The small size of the bodies (Phobos is 27 � 22 � 18 km(Archinal et al., 2011) and Deimos is 15 � 12 � 11 km(Murchie et al., 2015)) may be explained by a bombard-ment of planetesimals sourced from the vicinity of Jupiter(Safronov et al., 1986). In this accretionary model, themoons are composed of material that is similar to that ofMars, and are enhanced in components that were broughtin with the last heliocentric objects acquired by Mars(Burns, 1992). This model avoids the dynamical challengesassociated with the capture model.

Finally, it is possible that the martian satellites are theresult of a giant impact (Craddock, 2011; Citron et al.,2015; Rosenblatt et al., 2016). In this model, a planetesimalimpacted Mars, resulting in the formation of an accretiondisk around Mars. Material may have condensed and dis-sipated beyond the Roche limit, forming the satellites dueto gravity instabilities within the disk. In this scenario,the moons are loosely aggregated material from the accre-tion disk and do not contain any volatile elements(Craddock, 2011; Citron et al., 2015; Rosenblatt et al.,2016).

Either of these major models offers the opportunity toexplore interesting and important terrain, whether it bepieces of the outer solar system (e.g., Fraeman et al.,2012, 2014; Pajola et al., 2013; Pieters et al., 2014), buildingblocks of Mars (e.g., Burns, 1992; Rosenblatt andCharnoz, 2012; Peale and Canup, 2015), or materialderived from a later giant impact with Mars, such as mar-tian crustal or mantle composition (Craddock, 2011;Citron et al., 2015; Rosenblatt et al., 2016).

Different formation models imply different compositionsfor Phobos and Deimos (Table 1). If the bodies formed viaco-accretion, then they are expected to have moderatewater content of �1 wt% (Sarafian et al., 2014) and lowcarbon content (Murchie et al., 2014) with a compositionsimilar to bulk Mars (Wanke and Dreibus, 1988). Simi-larly, products from a martian impact scenario areexpected to have low water contents of �0.015 wt%(Truong and Lee, 2017) and low carbon values (Murchieet al., 2014) with a composition of evolved martian crustor mantle, similar to SNC meteorites (McSween et al.,2009). In contrast, captured carbonaceous asteroids areexpected to have relatively high (2–60 wt%) water contents(Murchie et al., 2014; Lee et al., 2017) and high carboncontents (Murchie et al., 2014). Thus, if the origin of themoons is determined to be capture, then the resourcepotential is strongest.

The volatile content, and by association the origin, ofPhobos and Deimos are major knowledge gaps in plane-tary protection for future human missions. For example,Phobos and Deimos may contain materials that interactedwith liquid water if they were derived from martian

near-surface material (e.g., Lee and Lorber, 2015). Addi-tionally, impacts can result in the transfer of materials fromMars to the moons (e.g., Ramsley and Head, 2013a).Therefore, both the mode of origin and material transferprocesses have important implications for the planetaryprotection of Phobos and Deimos (Melosh et al., 2012).

The most critical knowledge gap that exists in Phobosand Deimos science is the origin of these moons. The onlyway to unequivocally resolve whether Phobos or Deimosare captured bodies, are native to the martian system,and differ in origin, is to send a scientific mission that willcharacterize the moons in unprecedented detail and returnsamples. Resolving the origin of the martian satellites is thetop priority and may provide insight into the capture andevolution of asteroids or the formation and evolution ofterrestrial planets.

2.2. Composition

The reflectance and spectral slope of Phobos and Dei-mos are similar to carbon-rich CI and CM carbonaceouschondrites (e.g., Hiroi et al., 2003). Although they sharea similar spectral continuum slope to that of Mars crustalmaterial, Phobos and Deimos have spectral reflectancesthat are a factor of two darker than even the most space-weathered Mars crustal material, suggesting this is not agood spectral analog (Pajola et al., 2013). Deimos is char-acterized by a spectrally homogeneous redder unit, whilePhobos has two spectral units, a redder and a bluer unit.

Spectra acquired by the Compact Reconnaissance Imag-ing Spectrometer for Mars (CRISM) instrument onboardthe Mars Reconnaissance Orbiter (MRO) reveal a broad,shallow absorption centered at �0.65 lm on Deimos andon the redder unit on Phobos (Fraeman et al., 2014), whichcorrelates with the visible color observed by both CRISMand the High Resolution Imaging Science Experiment(HiRISE) (Murchie et al., 2015). Both the bluer (althoughmore weakly) and redder units on the satellites show a dis-tinct absorption �2.8 lm due to OH (Clark et al., 1990).These absorptions, the reflectance, and the spectral slopecharacteristics are consistent with a composition that isrich in phyllosilicate and that has been desiccated of H2O(Murchie et al., 2015). Phyllosilicate-rich CM carbona-ceous chondrites exhibit absorption features at �2.8 lmattributable to OH and at �0.7 lm due to Fe in phyllosil-icate (Cloutis et al., 2011; Takir et al., 2013). Because theFe feature observable in spectra of Phobos and Deimos iscentered near �0.65 lm, it has been suggested that Fe-bearing nontronite, a hydrous clay, is responsible and thatthe moons contain primitive material (Fraeman et al.,2014). Spectral mapping has also been done using thermalinfrared spectra. Glotch et al. (2015) found evidence forbound water and carbonate on the surface of Phobos, con-sistent with the mineralogy of D-type asteroids observed inthe Thermal Emission Spectrometer (TES) spectra. Alter-native models of the origins of the moons involve anhy-drous minerals. For example, space-weathering processes

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2177

produce microphase and nanophase Fe (e.g., Pieters et al.,2000), and Rayleigh scattering from these particlesmay then cause curvature in the spectra near 0.65 lm(Clark et al., 2012).

Determining the composition of the moons will place aprimary constraint on the origin of these bodies. High-resolution spectral measurements and multiple targets ofsample return are of top priority in characterizing themineralogy and volatile content of the satellites, andshould guide future scientific exploration of Phobos andDeimos.

2.3. Geology

Both Phobos and Deimos are characterized by low den-sities and low albedos, and have spectra consistent withthose of primitive outer solar system bodies (e.g., Bellet al., 1993). Both bodies also have albedo streaks on theinner and outer slopes of crater rims suggesting large-scale downslope movement. On Phobos, these streaks areinterpreted as talus and mass wasting mobilized in the geo-logically recent past (Basilevsky et al., 2014; Shi et al.,2016). Furthermore, thick (hundreds of m- to km-size)mounds on the slopes and floors of craters are observed,and are interpreted as landslide deposits (Basilevskyet al., 2014). Despite the micro-gravity at the surface (sur-face gravity on Phobos and Deimos is 0.0057 m/s2 and0.003 m/s2, respectively), impact-induced seismic shakingand surface movement produced by diurnal temperaturechanges can cause material to migrate along-slope(Basilevsky et al., 2014). Modeling of the formation ofthe albedo markings on Deimos suggests that vertical mix-ing, weathering, and creep, are important processes in themotion of material downslope, while impact gardeningcontributes very little (Thomas et al., 1996). A robotic mis-sion to the surface is critical to measure the efficiency andfrequency of material transport for the assessment ofinstrument and crew safety for surface operations on themoons.

Interestingly, despite these similarities, the regolith ofDeimos may be unique because its surface shows conspic-uous albedo patterns and is distinctively smooth (Thomaset al., 1996; Thomas et al., 2011). Understanding the rela-tionship between Phobos and Deimos is another key ques-tion remaining today, which can partly be elucidated bydetermining the origin of each moon. For example, if themoons co-accreted with Mars, then they are composition-ally related, however if the moons are captured bodies,then they may be unrelated or be fragments of a single par-ent body that was disrupted during capture (Singer, 2007;Rosenblatt, 2011). Additionally, because the moons mayhave been in closer orbits originally, regolith exchangebetween the two bodies may have occurred via a hypothe-sized dust belt (Soter, 1971) and thus sample returns ofboth moons are required to test whether the redder unitof Phobos could be derived from Deimos (Murchie et al.,2015). Understanding such differences is essential to

characterizing the formation and evolution of small bodiesin the solar system.

2.3.1. Grooves on Phobos

Parallel grooves exist on the surface of Phobos (Fig. 1)and are typically �100 to 200 m wide and several km long,defined by chains of coalescing pits of nearly the samediameter, grooves with scalloped margins, and grooveswith roughly linear margins (e.g., Thomas et al., 1979;Basilevsky et al., 2014; Murray and Heggie, 2014). Severalfamilies of features are defined by intersecting systems thathave approximately the same orientation (Thomas et al.,1979). Grooves are continuous features; often observedcutting through large craters and rims with no gaps, andeven intersecting other families of grooves withoutdisplaying lateral offset at intersections (Basilevsky et al.,2014).

Many formation models exist to explain the origin ofthese unique features. One possible explanation is thatthe grooves formed as fractures or faults due to tidal forces(e.g., Soter and Harris, 1977; Weidenschilling, 1979;Dobrovolskis, 1982; Hurford et al., 2016) or drag forcesupon capture (Pollack and Burns, 1977; Thomas et al.,1979). This explanation is consistent with the linear mor-phology of the grooves and the cross-cutting relationshipswith relatively older craters (Basilevsky et al., 2014). BruckSyal et al. (2016), however, modeled that the resulting dam-age patterns to Phobos from the Stickney-forming impactare unrelated to the grooved terrain.

Alternatively, it is possible that these grooves wereformed as chains of secondary impacts from Mars ejecta(e.g., Murray et al., 1992, 2014; Murray and Iliffe, 2011).In this model, each family of grooves represents a singleimpact event on Mars. The zone of avoidance on the trail-ing hemisphere of Phobos is explained by the velocity ofPhobos’ forward motion exceeding the ejecta velocity.However, Ramsley and Head (2013b) plotted precise Kep-lerian orbits for Mars ejecta to test this hypothesis andfound that this hypothesis is not consistent with Marsejecta emplacement models and observations. Specifically,they found that to emplace the observed grooves, grid pat-terns of (often tens of thousands of) ejecta fragments mustbe produced with nearly identical diameters, launched withvirtually zero rates of dispersion into fixed patterns ofarrays, which is statistically unlikely to occur. Additionally,they calculated that the volume of Mars ejecta that hasintersected Phobos is at least three orders of magnitude lessthan what is required to produce the grooves (Ramsley andHead, 2013a,b, 2014). Furthermore, such groove-formingimpacts would produce �25 to 50 m of new regolith onPhobos, burying features that are clearly observable today,such as spectral units and boulders (Ramsley and Head,2014). Other small bodies in the solar system have similargroove-like features, but lack a nearby body for crater-forming material to originate (Basilevsky et al., 2014).

Similarly, other researchers have suggested that thegrooves formed as chains of secondary impact craters or

2178 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

ejecta from Stickney (e.g., Veverka and Duxbury, 1977;Head and Cintala, 1979; Davis et al., 1980).

Finally, it is possible that the grooves are the remnanttracks of rolling and bouncing boulders (Head andWilson, 2011; Bruck Syal et al., 2016). Ejecta clasts mayhave slid, rolled, and bounced at distances comparable tothe observed groove lengths, crushing the regolith andpushing it aside as they moved (Head and Wilson, 2011).There are weaknesses to this model, namely that (1) thereis a distinct lack of boulders found at groove terminationsand (2) there is a lack of gaps where grooves cross positivetopography, which are expected to occur after a bouldercrosses a positive landform and bounces (Basilevskyet al., 2014). However, degradation of boulders since theformation of the grooves may have destroyed the blocks(Basilevsky et al., 2014). Also, the velocities of blocksejected during the formation of Phobos’ Stickney craterpermit the formation of continuous grooves across craterinteriors and rims (Wilson and Head, 2015).

In summary, none of the proposed formation mecha-nisms can suitably explain all features of the grooves.Important remaining strategic knowledge gaps (SKGs)are whether the origin and nature of the grooves on Pho-bos, and on other small bodies, are of the same genesis.Furthermore, what prevents Deimos from having these lin-ear features? A mission to Phobos can elucidate these enig-matic features by specifically testing these proposedmechanisms.

Different groove evolution scenarios may influencetraversibility constraints on Phobos for both robotic andhuman exploration. For example, a highly fractured sur-face may introduce stability constraints, and if groove for-mation is still active today, then this creates an active safetyhazard. Secondary impacts create chains of craters withdifficult slopes to navigate. Finally, remnant tracks mayhave rogue boulder and ejecta clasts in the terrain creatingnavigational obstacles. Overall, each model presents itsown set of navigational obstacles as explorational assetsapproach a set of grooves, but none preclude exploration.

2.4. Age

Another fundamental question regarding the moons ofMars is the age of the bodies. Stickney crater (Fig. 1) isthe largest crater on Phobos (�9 km in diameter).Schmedemann et al. (2014) derived crater production andchronology functions for two end-member scenarios ofthe dynamical evolution of Phobos: Case A assumes thatPhobos has been in its current orbit around Mars sinceits formation and Case B assumes a recent capture withan impact history of a typical Main Belt Asteroid. Theauthors suggest that (1) Phobos formed or experienced amajor collision at 4.3 Ga (Case A) or 3.5 Ga (Case B),(2) Stickney crater formed at �4.2 Ga (Case A) or 2.6 Ga(Case B), and (3) the grooves formed at �3.1 to 3.8 Ga(Case A) or 44–340 Ma (Case B) (Schmedemann et al.,2014). More recent work, however, suggests that many

craters on the surface of Phobos are the result of a sec-ondary impact spike produced from the Stickney-formingimpact ejecta, and that the surface age of Phobos maynot be derived accurately by counting such craters(Ramsley and Head, 2015, 2017). An alternative age forStickney crater of [0.5 Ga was derived by Ramsley andHead (2015, 2017) using the evidence of boulders and ejectablocks from Stickney (Thomas et al., 2000), boulderdestruction rates (Basilevsky et al., 2013, 2015), and spaceweathering rates of Phobos regolith (Pieters et al., 2014).An essential objective of a Phobos mission should beresolving the age paradox of Stickney crater, given thatage estimates for this feature span �3 Ga. Consequentially,placing absolute ages on different features and materials onthe surface of Phobos or Deimos will provide insight intothe age, formation, and origin of these bodies.

3. Exploration architecture

We developed a framework of scientific questions toassess prior to sending humans to the moons of Mars. Inorder to determine the role of Phobos and/or Deimos inthe future exploration of Mars, their near-surface physicaland chemical characteristics must be understood. To gainan understanding of these properties, the key question thatinitially drives this framework is: What is the origin of Pho-bos and Deimos? Several formation models have been pro-posed, as discussed in Section 2.1. Each formation modelhas distinct implications for the utility of Phobos and Dei-mos in the human exploration of Mars (Table 1), where acaptured body would provide the most resources.

The SKGs can be broadly divided into three categories:(1) the composition, (2) overall surface characteristics, and(3) geophysical properties of Phobos and Deimos (Table 2).Compositional characterization will include orbital-scalemapping of compositional units in the VIS, IR, and near-IR, sample return of redder and bluer spectral units, andidentification and distribution of potential resources. Sur-face geologic characterization will include global high-resolution imaging and topographic characterization.Finally, geophysical characterization will include definingphysical properties of regolith, global mapping of thenear-surface gravity field, and mapping the internal struc-ture of both bodies.

For human exploration of the moons, answers to SKGcategories 1 and 2 (Table 2) feed forward into in situ

resource utilization (ISRU) capabilities. The compositionand geological characteristics have critical implicationsfor sample return sites and ISRU targeting. If Phobos isa spectrally bluer body that is covered by a redder mantledue to space weathering or dust, then one sampling site issufficient to acquire samples from both units (Basilevskyet al., 2014). If Phobos is heterogeneous, however, thenmultiple sampling sites are required to understand the his-tory and composition of the moon (Basilevsky et al., 2014).Furthermore, the spatial scale and distribution of theseunits must be well characterized before any sample

Table 2SKGs for Phobos and Deimos exploration.

Category 1:Composition

� What is the origin of each body?� What factors contribute to the low density of Phobos (i.e. water ice or pore space)?� How thick is the regolith?� What makes up the distinctive blue and red units on Phobos?� What is the mineralogy and petrology of Phobos and Deimos?� What is the quantitative abundance of Mars ejecta in regolith?� What is the quantitative abundance of H2O- and OH-bearing phases?� What is the quantitative abundance of organic, C-bearing phases?

Category 2:Surface characteristics

� What are the regolith mechanical properties?� What is the age of Stickney crater and how does it relate to the crater size-frequency distribution?� What are the origin, distribution, and age of the boulders on Phobos?� What is the origin of the grooves on Phobos?

Category 3:Geophysical properties

� What is the internal structure of Phobos and Deimos?� What factors contribute to the low density of Phobos?� What is the radiation dose at each body?� What is the global shape and rotational state of each body?� What is the gravitational field at each body?

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2179

acquisition instrumentation is deployed, which is a majorgoal in the orbital phase of our exploration framework(Fig. 2, Stage I).

SKG categories 2 and 3 (Table 2) feed forward intotraversing capabilities. Understanding the geological andgeophysical properties has specific implications for detailedtraversibility maps for future robotic and human missions.Details of surface geology, specifically regolith properties(e.g., dielectric constant and porosity), surface structure,compactness, slope measurements, and landslide hazards,are critical for the characterization of environmentalhazards.

We have devised a two-mission-architecture (Fig. 2) toaddress our science-driven questions with the added benefitof providing key information for later missions involvinghuman exploration.

3.1. Robotic exploration of Phobos and Deimos

The main science goal of this mission is to characterizethe surface of Phobos and Deimos using both orbitalspacecraft instruments and a smaller landed spacecraft,and to provide returned samples for confirmation of thein situ compositional analysis, similar to the objectives ofthe attempted Russian sample return mission Fobos-Grunt (Marov et al., 2004). Our architecture is patternedoff of the Hayabusa 2 mission (Tsuda et al., 2013) in bothspacecraft infrastructure and mission operations, andinvolves an initial orbital mapping stage, detachment ofthe landed assets and operation throughout their lifetime,and is followed by spacecraft sample acquisition and sam-ple return to Earth. The robotic mission is composed oftwo stages (Fig. 2, Stages I–II).

3.1.1. Stage I: orbital exploration of Phobos and DeimosThe orbital phase of the robotic exploration architecture

(Fig. 2, Stage I) is designed to address the major SKG: the

origin of the moons. As discussed in Section 2.1, there aremany origin hypotheses that have been proposed to explainthe moons of Mars, and each has distinct implications forthe composition and utility of the moons (Table 1). Theorbiter includes a laser altimeter, high-resolution camera,spectrometer, and gravity ranging system. The tangiblescience products of the orbital mission include global topo-graphic coverage of Phobos and Deimos generated by orbi-tal stereo imaging and laser altimetry, global imaging ofPhobos and Deimos at sub-meter resolution, global com-positional mapping of Phobos and Deimos via orbital spec-troscopy, and a global map of the near-surface gravityfield.

The products of Stage I are designed to address SKGcategories 1 and 2: the compositional and surface charac-terization of Phobos and Deimos (Table 2). There aremany feed-forward benefits of this mission. First, we willproduce detailed maps for future mission planning for bothrobotic and human exploration, which are essential whenassessing the safety of the terrain, but also when selectingthe most scientifically interesting sites for surface explo-ration. Composition maps will play a crucial role in tra-verse planning for sample collection and resourcetargeting. Second, a compositional understanding of sur-face materials and properties provides information ontraversibility and sample collection priorities. Third, weanalyze the potential resource availability and distribution.

High-resolution spectrometers that provide diagnosticelemental abundances and mineralogy of fresh, non-space-weathered surfaces are critical in resolving the abun-dances of major and minor elements and the compositionsof specific units. This provides vital evidence required toconstrain the origin of the moons. Furthermore, high-resolution topographic measurements and color observa-tions of the grooves on Phobos can test different grooveformation models, because each model makes different pre-dictions for the topography of groove rims and floors and

2180 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

the stratigraphic relationships between groove floors andrim materials (Murchie et al., 2015). In addition, determin-ing the density to find the internal macroporosity and vol-ume of the bodies is important in constraining the deepstructure of the moons and possible deep water-ice inven-tory (Murchie et al., 2015).

3.1.2. Stage II: landed exploration of Phobos/DeimosThe objective of Stage II is to characterize the structure

of Phobos, and to provide an enhanced sample suite for thecompositional diversity of Phobos by utilizing a robotic

Fig. 2. Flowchart illustrating the scientific exploration framework for Phohighlighted in green, and human exploration phases are highlighted in yellow. Rdetermined or that Phobos/Deimos do not offer a particular resource along theDeimos is a weighted algorithm of many factors. Multiple ‘‘no’s” do not necesthe references to color in this figure legend, the reader is referred to the web v

lander that will also have roving capabilities (Fig. 2,Stage II).

The landed mission is designed to characterize the sur-face at higher spatial resolution to prepare for futurerobotic and human surface operations. The landed instru-mentation has mobility, assembled onto ‘‘hedgehogrobots,” which are specifically designed to operate in low-gravity conditions and rough surfaces (Pavone et al.,2013). These cube-shaped robots can hop and tumble onthe surface, independent of one another, while also safelyhousing instruments (Pavone et al., 2013). The landed

bos/Deimos. Orbital phases are highlighted in blue, landed phases areed ‘‘no” boxes suggest that the answer to the specified question cannot bepath in human martian exploration. The exploration to Mars via Phobos/sarily warrant a halt in the exploration framework. (For interpretation ofersion of this article.)

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2181

components allow for in situ hyperspectral imaging, visibleimaging, measurements of the magnetic field, and acquisi-tion of infrared and thermal data, from a wide range oflocations. Finally, Stage II will also employ sample collec-tion in order to quantitatively confirm spacecraft/landerobservations and to produce samples for laboratory analy-sis on Earth, for the curation of samples for long-term stor-age and global community retrieval. Terrestriallaboratories will allow for the comprehensive analyses ofthe mineralogical, elemental, and isotopic composition ofsamples at a level of detail that is not possible with roboticrendezvous.

The lander payload is specifically designed to investigatethe electrical, mechanical, and porosity properties of theregolith, characterize the internal structure of the body inorder to quantify how competent the body is, and to returna substantial sample collection of several grams. Theseproducts seek primarily to address SKG categories 1 and3: the compositional and geophysical characterization(Table 2). The data obtained from this stage, informedby the previous stage, should provide an answer to ourdriving science question of ‘‘What is the origin of Phobosand Deimos?”. The specification and scope of subsequentlanded missions depend on the results from the priormission.

The feed-forward benefits of Stage II are related toassessing the safety of the martian moon system for humanexploration and to appraising the value that the moons canoffer in radiation/micrometeorite/dust protection, technol-ogy development, and resources. We will produce adetailed characterization of surface conditions includingthe radiation environment to determine the necessaryradiation-protection shelter required for humans andinstruments. The orbital debris environment must be char-acterized in order to assess possible hazards from regolithdust. We will gain experience working and roving in themicro-gravity environment in preparation for using real-time computer-assisted traverses to guide humans aroundthe surface in later missions. Finally, the distribution andabundance of important resources for supporting humanoperations or providing sources for fuel, such as hydratedminerals and carbon, will be characterized at highresolution.

Note that in Fig. 2, beginning in Stage II, each of thequestions terminate in the selection of either ‘‘yes,” tocontinue along the mission framework, or ‘‘no,” to endthat route of inquiry. We do not suggest that a single‘‘no” is enough to preclude Phobos or Deimos from aid-ing in the exploration of Mars, but rather that that partic-ular question failed to provide a potential merit in doingso. Thus, multiple ‘‘no’s” should not necessarily prevent ahuman exploration program to the moons of Mars. Theanswer to this driving question is a weighted algorithmof each of the termini where the most important factoris determining to what extent Phobos/Deimos can aid inthe future exploration of Mars. The flowchart alsoincludes a cyclic option for additional, independent

landed missions as needed to address the defined SKGs(Fig. 2, dashed lines).

3.2. Stage III: assess role of human exploration of Phobos/

Deimos

The objective of Stage III (Fig. 2) is to determinewhether humans can successfully and safely operate onthe surface of the moons using Stage I–II science. StageIII is an exploratory phase that is not an additional mis-sion, but in which we re-assess the results of the precedingrobotic exploration of the moons (Stages I–II). Specifically,we seek to determine if the debris, radiation, gravity, androtational environments are understood in sufficient detailto address astronaut safety concerns (Fig. 2, Stage III).

Before sending astronauts to the moons of Mars, wemust understand the influence of micro-gravity on astro-naut health. Long-duration spaceflight to the martian sys-tem and a basecamp in a micro-gravity environment suchas Phobos or Deimos present several health risks due tothe hostile space environment, including loss of bone mass,muscle atrophy, cardiac dysrhythmias, and altered orienta-tion (Blaber et al., 2010).

These three topics (the debris, radiation, and gravita-tional environment) are essential in characterizing the envi-ronment for human safety. The following two topics,gravitational and rotational states, are imperative for oper-ational success.

Characterizing the debris environment (Fig. 2, Stage III)is critical in identifying any potential hazards in the near-surface environment for both human explorers and infras-tructure. Ejecta from impacts into Mars typically impactsthe surface of Phobos with a velocity of 2–3 km/s, produc-ing ejecta from secondary impacts on Phobos with veloci-ties <800 m/s, and subsequently 95–99% of the Phobosejecta remains in orbit around Mars (Ramsley and Head,2013a). In contrast, ejecta that is produced by primaryimpacts on Phobos has higher velocities, and a smaller pro-portion of material re-impacts or is gravitationally cap-tured (Ramsley and Head, 2015). Understanding thefrequency of impacts as well as the dust and debris trappedin the orbital environment is an important aspect of assess-ing the safety of human operations on the moons.

In addition to dust, it is important to quantify the radi-ation environment and the effects of long-term exposure toboth humans and instrumentation (Fig. 2, Stage III). Thelocation of the base may help reduce radiation exposure.For example, surrounding terrain may help provide protec-tion, such as if the surface operation base is settled withinStickney crater on Phobos. In this configuration, nearlyone hemisphere of radiation is blocked by Mars and Pho-bos (Mueller and Metzger, 2015). In addition, a regolithof sufficient thickness (�2 to 5 m) may provide an adequateradiation shielding effect to stop secondary radiation,caused by cosmic radiation shattering nuclei in the targetmaterial (Mueller and Metzger, 2015). In this case, it may

2182 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

be useful to coat the habitat in regolith material. Alterna-tively, water may provide yet the best shielding material.

The gravitational and rotational states of the body arecritical to quantify and map before arrival for the opera-tional design of instrumentation, equipment deployment,and traversibility maps (Fig. 2, Stage III). Small bodiescharacteristically have non-intuitive vectors for surfaceacceleration, derived from irregular shapes, rapid spin,low mass, and tidal effects (Murchie et al., 2015). These fac-tors can create a particularly challenging environment forhumans to navigate, thus pre-calculated traverses basedon the potential energies over the surface are necessaryfor operational success. Although the micro-gravity envi-ronment will present operational challenges to the crew,it also contributes to exploration by promoting moregentle-walled craters compared to lunar highland cratersof the same size (Basilevsky et al., 2014). Shallower slopesprovide safer terrains for building camp and for roboticand human traversing. If these fields are well characterizedand conditions are safe for human explorers, proceed to thefinal framework phase, Stage IV.

It is possible that planned upcoming missions (Table 3)can fulfill Stages I–III of the framework by characterizingthe origin and thus the distinct utility of the moons, as wellas the safety and operational constraints of the martiansystem. These missions (Table 3) have similar payloads towhat we have suggested in Section 3. With the tentativedates planned for sample returns to Earth in the late2020’s (Table 3), it is possible that the framework permitshuman missions to the moons in the 2030’s.

3.3. Stage IV: human exploration of Phobos/Deimos

The final stage of the mission framework consists of asuite of tasks that human explorers can accomplish onthe moons of Mars (Fig. 2, Stage IV). Early design refer-ence scenarios were constructed for manned missions toPhobos and Deimos by O’Leary (1985) and Aaron(1988), and continue to be developed by the Mars MoonsTeam at Johnson Space Center (Gernhardt, 2015, 2016).These design concepts suggest that humans could accom-plish substantial science at Phobos while establishing early

Table 3Planned future missions to Phobos.

Mission Agency Objectives

Mars Moon eXplorationa JAXA – Determi– Understa

Phobos sample returnb ESA-ROSCOSMOS – Study bu– Return s– Demonsfor Mars

a Fujimoto et al. (2017).b Chalex et al. (2014).

leadership in the martian system and continuing program-matic exploration (O’Leary, 1985; Aaron, 1988;Gernhardt, 2015, 2016).

We first and foremost note that humans can accomplishscience (Fig. 2, Stage IV). Humans can greatly increase theefficiency in sample collection and sample return capacityby drilling for and collecting deep geologic samples(Crawford, 2012). Humans can also make observations ofother targets, including Mars and the neighboring moon.They can also sample large ejecta blocks, and identifyand sample exogenic materials (Lee et al., 2017). The scien-tific output achieved by humans is substantial and humanexplorers can greatly assist in resolving remaining SKGs,and additional SKGs that develop as the mission explo-ration framework proceeds and new data are collected.

In addition, humans can set-up and operate an ISRUextraction base (Fig. 2, Stage IV) to harness regolith andsilicate minerals (which host, e.g., silicon, iron, aluminum)for building materials, carbon for fuel, or possible hydratedminerals for water or propellant (Nichols, 1993). Addition-ally, sunlight can be harnessed as a source of solar energy.A variety of derivate resources can be produced from avail-able materials including plastics made from carbon andmetals and ceramics refined from silicate minerals, and car-bonaceous materials can be processed for carbon, hydro-gen, and nitrogen in refueling and life support operations(O’Leary, 1985). If the moons are determined to be of acaptured origin, then we expect these materials to be muchmore abundant (Table 1), and thus the moons offer moreresource potential in aiding martian operations.

Humans can also increase the potential for large-scaleexploratory activities by deploying large-scale and complexequipment and can further the development of a space-based infrastructure by providing routine and emergencymaintenance of the complex equipment and habitat(Crawford, 2012). Finally, an important manufacturingtask for human crews may include the creation of heatshields and extraction of metals to 3D-print spare parts(Mueller and Metzger, 2015).

Furthermore, humans can demonstrate the capability tooperate in a micro-gravity environment, which has not yetbeen experienced by humans (Fig. 2, Stage IV). Humans

Timeline

ne origin of moonsnd processes in circum-martian environment

Launch 2024Mars arrival 2025Earth arrival 2029

lk characteristics of martian moonsurface samples from Phobostrate and mature technologies requiredsample return missions

Launch 2024Mars arrival 2025Earth arrival 2027

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2183

can assist in the incremental development of technologiestested in the martian system as they are challenged by anew environment.

On the surface of Phobos or Deimos, humans can alsobuild and maintain a telecommunications base (Fig. 2,Stage IV). Early Mars missions may not land on Marsdue to the high financial cost of, and need for, improvedEntry, Descent, and Landing (EDL) technologies. Ahuman mission to Phobos can assist with the emplacementof a simple Mars telecommunications network, either witha series of satellites providing low-Mars-orbit transmissionand reception relay capabilities on different orbits, or bythe emplacement of aerostationary satellite capabilities.Alternatively, astronauts on Phobos can oversee robotson the martian surface in preparation for later human land-ings on Mars. These robotic operations may include min-ing activities, creating propellants for Mars ascent, andbuilding landing pads – all productive steps forward inNASA’s evolvable Mars campaign. We suggest that thiswill not only be significant in the preparation for humanarrival on Mars, but also critically important in the contin-uation of the martian program.

Finally, humans may construct radiation protectionfrom surface regolith with ISRU equipment (Fig. 2, StageIV). This is a vital task that can help lower the radiationdose for both human crewmembers and electronics, as dis-cussed in Section 3.2.

4. Discussion

Both Phobos and Deimos are interesting explorationdestinations that warrant further exploration on theirown merit. The scientific exploration architecture discussedhere is designed to explore both bodies robotically in orbi-tal and lander phases (Stages I–II). When consideringhuman exploration, however, we also consider factors ofhuman safety and infrastructure safety and reliability.

One major constraint for human exploration of the mar-tian moons is that both bodies have low escape velocities;the escape velocity of Phobos is 11.39 m/s while that onDeimos is only 5.56 m/s (Gernhardt, 2015, 2016). Suchan extreme micro-gravity environment presents a majorchallenge to field astronauts, who can potentially launchthemselves off the body without expelling substantialenergy. For example, a suited crewmember who jumps ver-tically with a 2 m/s take-off velocity reaches a verticalheight of 1.2 m on the Moon, and reaches back down tothe surface in 2.5 s (Gernhardt, 2015, 2016). On Phobos,however, this same suited crewmember would reach350.9 m above the body, and not return to the surfacefor 11.7 min (Gernhardt, 2015, 2016). And on Deimos,the trip would last for 22.2 min as the crewmember peakedat 666.7 m (Gernhardt, 2015, 2016).

Another issue for human exploration is extendedexposure to radiation. Stickney crater on Phobos, how-ever, has the potential to provide additional radiationshielding for instruments and crews. Setting up a base

within the crater can potentially block nearly one hemi-sphere of radiation (Mueller and Metzger, 2015), whichcould amount to a substantial reduction in radiationdose over time.

While both Phobos and Deimos warrant further scien-tific exploration, we conclude that Phobos is the mostviable human exploration target because of its higher sur-face gravity, greater escape velocity, and more potentialradiation shielding. However, the scientific explorationframework presented here (Fig. 2) is designed to exploreboth targets robotically in order to address the scientificallydriven SKGs (Table 2) defined for the moons of Mars.Both moons are explored to determine the nature and ori-gin of these enigmatic bodies. During the human explo-ration phase, observational periods of Deimos take placewith high-resolution equipment to further characterizethe fellow moon of Mars (Fig. 2, Stage IV).

Sending humans to the surface of Phobos presents sev-eral advantages over going directly to Mars. Overall, thebulk cost of sending humans to the surface of Phobos isconsiderably less than it is to send astronauts to the surfaceof Mars; Phobos can be reached much sooner than Mars,and for a far lower monetary and human risk cost.Through this exploration science framework, we haveshown that the moons of Mars are exciting as both scien-tific and exploration destinations. The regolith of Phobosmay offer the opportunity to study small bodies/asteroids,martian material, and volatiles on a single surface, whileanswering a variety of SKGs (Table 2). It has been pro-posed that samples that are collected robotically on Marscan be cached on the surface of Phobos for humans toretrieve later (Stooke, 2014; Lee et al., 2017).

As exploration targets, Phobos and Deimos offer stableorbital platforms for Mars observations and communica-tions. Furthermore, the journey to these moons requiresa minimal delta-V from the Earth, making them a favor-able destination in comparison to the more challengingMars (O’Leary, 1985). Finally, a successful human pro-gram to Phobos undoubtedly serves as a catalyst for get-ting humans to Mars.

If no humans are sent to Mars before we are preparedfor a surface landing on the planet, then decades mayelapse before we are prepared for the journey. Phobos,however, presents key programmatic advantages in sendinghumans to the martian system as intermediate steps toensure continuity in the martian program: (1) Phobos is atechnically achievable martian target, requiring only low-cost near-term development of lunar and/or ISS systems(Lee et al., 2005). (2) A journey to Phobos reduces riskfor a Mars landed mission through a stepwise programand technology demonstrations (Lee et al., 2005). (3) Pho-bos exploration enables the achievements of an excitingand high-science return mission in the near term to themartian system, maintaining programmatic focus and con-tinued public support (Lee et al., 2005).

Thus, Phobos and Deimos present strong cases tosupport the future exploration of Mars. Their roles in

2184 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

supporting the future exploration of Mars are largely dri-ven by: (1) the strong science return that visiting these smallbodies guarantees, specifically in addressing their originand composition (Table 2), and most importantly (2) a suc-cessful human program to Phobos is a catalyst for a suc-cessful human mission to Mars and ensures a steadycadence of meaningful, near-term missions in the martiansystem (Lee et al., 2005).

5. Conclusions

There are still many unknowns surrounding the natureand origin of the moons of Mars, and further roboticmissions are required to characterize these bodies. Thesemissions may also reveal to what extent Phobos and Dei-mos can enhance the martian exploration program. Herewe have developed an architecture for what must beassessed before we can send humans to the surface ofthese moons. Programmatic logic may dictate sendinghumans to Phobos or Deimos, in which case this missionframework is critical when structuring a human precur-sor mission. We believe that sending humans will alwaysbe beneficial based on the scientific output they canaccomplish.

Phobos and Deimos are positioned to support Mars sur-face operations. A Phobos mission establishes early leader-ship by sending humans to the martian system to explore,conduct resource surveys, and establish a scientific stationthat builds upon martian assets and can allow for samplereturn of both martian and Phobos regolith to Earth fordetailed analysis (Aaron, 1988). If the moons containhydrated minerals, mining may be a key strategy for afford-able Mars campaigns. While both moons are interestingexploration destinations, we conclude that Phobos is asafer target due to higher surface gravity, greater escapevelocity, and more potential radiation shielding. Further-more, Phobos may provide a suitable location for astro-nauts to maintain a telecommunications network. ISRUoperations may include the production of heat shieldsand 3D-printing spare parts. The exploration of Phobosis potentially an affordable and productive first-step towardsurface operations on Mars. Overall, Phobos is a scientifi-cally interesting destination that offers engineering, opera-tional, and public engagement benefits that mostimportantly would allow for the continuation and develop-ment of the Mars exploration campaign in the upcomingdecade.

The exploration framework laid out here seeks toaddress how Phobos and Deimos can aid in the futureexploration of Mars. Phobos and Deimos are potentiallyvaluable commodities, providing a wealth of science return,as well as telecommunications capabilities, resource utiliza-tion, radiation protection, transportation and operationsinfrastructure, and the continuity of the martian explo-ration program.

Acknowledgements

We gratefully acknowledge the NASA Solar SystemExploration Research Virtual Institute (SSERVI) for spon-soring the graduate seminar ‘‘Phobos and Deimos: TheMoons of Mars,” which motivated this work. We alsothank Dr. John Grunsfeld, former Associate Administratorfor the Science Mission Directorate and NASA Astronaut,and NASA Astronaut Mike Gernhardt for helpful discus-sions about this framework. We thank Ernesto Palombaand Peggy Ann Shea for editorial handing of the manu-script and two anonymous reviewers for insightful reviews.

References

Aaron, J., 1988. Beyond Earth’s Boundaries: Human Exploration of theSolar System in the 21st century. NASA Office of Exploration,Washington, D.C., p. 51.

Archinal, B.A., A’Hearn, M.F., Bowell, E., Conrad, A., Consolmagno, G.J., Courtin, R., Fukushima, T., Hestroffer, D., Hilton, J.L., Krasinsky,G.A., Neumann, G., Oberst, J., Seidelmann, P.K., Stooke, P., Tholen,D.J., Thomas, P.C., Williams, I.P., 2011. Report of the IAU WorkingGroup on cartographic coordinates and rotational elements: 2009.Celest. Mech. Dyn. Astr. 109, 101–135. https://doi.org/10.1007/s10569-010-9320-4.

Basilevsky, A.T., Head, J.W., Horz, F., 2013. Survival times of meter-sized boulders on the surface of the Moon. Planet. Space Sci. 89, 118–126. https://doi.org/10.1016/j.pss.2013.07.011.

Basilevsky, A.T., Head, J.W., Horz, F., Ramsley, K., 2015. Survival timesof meter-sized rock boulders on the surface of airless bodies. Planet.Space Sci. 117, 312–328. https://doi.org/10.1016/j.pss.2015.07.003.

Basilevsky, A.T., Lorenz, C.A., Shingareva, T.V., Head, J.W., Ramsley,K.R., Zubarev, A.E., 2014. The surface geology and geomorphologyof Phobos. Planet. Space Sci. 102, 95–118. https://doi.org/10.1016/j.pss.2014.04.013.

Bell, J., Fanale, F., Cruikshank, D.P., 1993. Chemical and physicalproperties of the martian satellites. In: Lewis, J. (Ed.), Resources ofNear-Earth Space. Univ. Arizona, Tucson, pp. 887–901.

Blaber, E., Marcal, H., Burns, B.P., 2010. Bioastronautics: the influence ofmicrogravity on astronaut health. Astrobiology 10, 463–473. https://doi.org/10.1089/ast.2009.0415.

Bruck Syal, M., Rovny, J., Owen, J.M., Miller, P.L., 2016. ExcavatingStickney crater at Phobos. Geophys. Res. Lett. 43, 10595–10601.https://doi.org/10.1002/2016GL070749.

Burns, J.A., 1978. The dynamical evolution and origin of the martianmoons. Vistas Astron. 22, 193–210. https://doi.org/10.1016/0083-6656(78)90015-6.

Burns, J.A., 1986. Some background about satellites. In: Burns, J.A.,Matthews, M.S. (Eds.), Satellites. University of Arizona Press.

Burns, J.A., 1992. Contradictory clues as to the origin of the martianmoons. In: Mars. University of Arizona Press, pp. 1283–1301.

Chalex, R., Ferracina, L., Romstedt, J., et al., 2014. CDF Study Report:Phobos Sample Return. ESA Science and Robotic ExplorationDirectorate. Advanced Studies and Technology Preparation Division,Paris, France, p. 254.

Citron, R.I., Genda, H., Ida, S., 2015. Formation of Phobos and Deimosvia a Giant Impact. Icarus 252, 334–338. https://doi.org/10.1016/j.icarus.2015.02.011.

Clark, R.N., Cruikshank, D.P., Jaumann, R., Brown, R.H., Stephan, K.,Dalle Ore, C.M., Eric Livo, K., Pearson, N., Curchin, J.M., Hoefen, T.M., Buratti, B.J., Filacchione, G., Baines, K.H., Nicholson, P.D.,2012. The surface composition of Iapetus: mapping results fromCassini VIMS. Icarus 218, 831–860. https://doi.org/10.1016/j.icarus.2012.01.008.

A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186 2185

Clark, R.N., King, T.V.V., Klejwa, M., Swayze, G.A., Vergo, N., 1990.High spectral resolution reflectance spectroscopy of minerals. J.Geophys. Res. 95, 12653–12680. https://doi.org/10.1029/JB095iB08p12653.

Cloutis, E.A., Hudon, P., Hiroi, T., Gaffey, M.J., Mann, P., 2011. Spectralreflectance properties of carbonaceous chondrites: 2. CM chondrites.Icarus 216, 309–346. https://doi.org/10.1016/j.icarus.2011.09.009.

Craddock, R.A., 2011. Are Phobos and Deimos the result of a giantimpact? Icarus 211, 1150–1161. https://doi.org/10.1016/j.icarus.2010.10.023.

Crawford, I., 2012. Dispelling the myth of robotic efficiency. Astron.Geophys. 53, 2.22–2.26. https://doi.org/10.1111/j.1468-4004.2012.53222.x.

Davis, D.R., Weidenschilling, S.J., Chapman, C.R., Greenberg, R.,Housen, K.R., 1980. Dynamical studies of Phobos and Deimos:groove origin and ejecta dynamics. Reports of planetary geologyprogram, 14–16.

Dobrovolskis, A.R., 1982. Internal stresses in Phobos and other triaxialbodies. Icarus 52, 136–148. https://doi.org/10.1016/0019-1035(82)90174-9.

Fraeman, A.A., Arvidson, R.E., Murchie, S.L., Rivkin, A., Choo, T.,Bibring, J.-P., Gondet, B., Humm, D., Kuzmin, R.O., Manaud, N.,Zabalueva, E.V., 2012. Analysis of disk-resolved OMEGA andCRISM spectral observations of Phobos and Deimos. J. Geophys.Res. 117, E00J15. https://doi.org/10.1029/2012JE004137.

Fraeman, A.A., Murchie, S.L., Arvidson, R.E., Clark, R.N., Morris, R.V., Rivkin, A.S., Vilas, F., 2014. Spectral absorptions on Phobos andDeimos in the visible/near infrared wavelengths and their composi-tional constraints. Icarus 229, 196–205. https://doi.org/10.1016/j.icarus.2013.11.021.

Fujimoto, M., 2017. ISAS Program for Small Body Exploration. NASAExplor. Sci. Forum, abstract NESF2017-102.

Gernhardt, M., 2015. Phobos as an exploration destination and base forMars exploration Phobos and Deimos: The Moons of Mars AcademicGraduate-Level Course. Invited Talk. Archived online https://sservi.-nasa.gov/event/planetary-evolution-phobos-and-deimos/.

Gernhardt, M., 2016. Human exploration of Phobos. In: Thirdinternational Conference on the Exploration of Phobos and Deimos.Invited talk. Archived online https://phobos-deimos.arc.nasa.gov/on-demand/.

Glotch, T., Edwards, C., Ebel, D., 2015. Spectral properties of Phobosfrom the Mars Global Surveyor thermal emission spectrometer:evidence for water and carbonate. Lunar Planet. Sci. 46, abstract 2587.

Hartmann, W.K., 1990. Additional evidence about an early intense flux ofC asteroids and the origin of Phobos. Icarus 87, 236–240. https://doi.org/10.1016/0019-1035(90)90033-6.

Head, J.W., Cintala, M.J., 1979. Grooves on Phobos: evidence forpossible secondary cratering origin. Reports of planetary geologyprogram, pp. 19–21.

Head, J.W., Dickson, J., Mustard, J., Milliken, R., Scott, D., Johnson, B.,Marchant, D., Levy, J., Kinch, K., Hvidbery, C., Forget, F., Boucher,D., Mikucki, J., Fastook, J., Klaus, K., 2015. Mars Human ScienceExploration and Resource Utilization: The Dichotomy BoundaryDeuteronilus Mensae Exploration Zone. First landing site/explorationzone workshop for human missions to the surface of Mars, abstract1033.

Head, J.W., Wilson, L., 2011. Origin of grooves on Phobos andcomparisons to the Moon. In: Proceedings of the 2nd Moscow SolarSystem Symposium, abstract 2MS3-MM-01.

Hiroi, T., Moroz, L.V., Shingareva, T.V., Basilevsky, A.T., Pieters, C.M.,2003. Effects of microsecond pulse laser irradiation on Vis-NIRreflectance spectrum of carbonaceous chondrite simulant: implicationsfor martian moons and primitive asteroids. Lunar Planet. Sci. 34,abstract 1324.

Hunten, D.M., 1979. Capture of Phobos and Deimos by photoatmo-spheric drag. Icarus 37, 113–123. https://doi.org/10.1016/0019-1035(79)90119-2.

Hurford, T.A., Asphaug, E., Spitale, J.N., Hemingway, D., Rhoden, A.R.,Henning, W.G., Bills, B.G., Kattenhorn, S.A., Walker, M., 2016. Tidaldisruption of Phobos as the cause of surface fractures. J. Geophys.Res. Planets 121, 1054–1065. https://doi.org/10.1002/2015JE004943.

Lee, P., Braham, S., Mungas, G., Silver, M., Thomas, P., West, M., 2005.Phobos: A Critical Link Between Moon and Mars Exploration. SpaceResources Roundtable VII, abstract 2049.

Lee, P., Lorber, K., 2015. Phobos and Deimos: Planetary ProtectionKnowledge Gaps for Human Missions. Planetary Protection Knowl-edge Gaps for Human Extraterrestrial Missions, abstract 1007.

Lee, P., Rosenblatt, P., Sims, M., Tagaki, Y., Truong, N., 2017. Phobosand Deimos: science and exploration objectives for humans. In: NASAExplor. Sci. Forum, abstract NESF2017-116.

Marov, M.Y., Avduevsky, V.S., Akim, E.L., Eneev, T.M., Kremnev, R.S.,Kulikov, S.D., Pichkhadze, K.M., Popov, G.A., Rogovsky, G.N.,2004. Phobos-Grunt: Russian sample return mission. Adv. Space Res.Mercury, Mars Saturn 33, 2276–2280. https://doi.org/10.1016/S0273-1177(03)00515-5.

McSween, H.Y., Taylor, G.J., Wyatt, M.B., 2009. Elemental compositionof the Martian crust. Science 324, 736–739. https://doi.org/10.1126/science.1165871.

Melosh, J., Conley, C., Kminek, G., Chappaz, L., 2012. Planetaryprotection status of Phobos and Deimos. In: Presented at the 39thCOSPAR Scientific Assembly, p. 1224.

Mueller, R., Metzger, P., 2015. Phobos-Deimos In Situ Resource Utiliza-tion (ISRU). Phobos and Deimos: The Moons of Mars AcademicGraduate-Level Course. Invited Talk. Archived online https://sservi.-nasa.gov/event/planetary-evolution-phobos-and-deimos/.

Murchie, S.L., Britt, D.T., Pieters, C.M., 2014. The value of Phobossample return. Planet. Space Sci. 102, 176–182. https://doi.org/10.1016/j.pss.2014.04.014.

Murchie, S.L., Thomas, P.C., Rivkin, A.S., Chabot, N.L., 2015. Phobosand Deimos. In: Asteroids IV, pp. 451–467. https://doi.org/10.2458/azu_uapress_9780816532131-ch024.

Murray, Iliffe, J.B., 2011. Morphological and geographical evidence forthe origin of Phobos’ grooves from HRSC Mars Express images.Martian Geomorphology 356, 21–41. https://doi.org/10.1144/SP356.3.

Murray, J.B., Heggie, D.C., 2014. Character and origin of Phobos’grooves. Planet. Space Sci. 102, 119–143. https://doi.org/10.1016/j.pss.2014.03.001.

Murray, J.B., Rothery, D.A., Thornhill, G., Muller, J.-P., Cook, T., Day,T., Iliffe, J.C., 1992. The origin of grooves and craters chains ofPhobos. Lunar Planet. Sci. 23, abstract 1472.

Nichols, C.R., 1993. Volatile products from carbonaceous asteroids. In:Resources of Near-Earth Space. University of Arizona Press, Tucson,pp. 543–568.

O’Leary, B., 1985. Phobos and Deimos (PhD): Concept for an earlyhuman mission for resources and science. In: Princeton/AIAAConference on Space Manufacturing, pp. 41–48.

Oberst, J., Zakharov, A., Schulz, R., 2014. Why study Phobos andDeimos? An introduction to the special issue. Planet. Space Sci. 102, 1.https://doi.org/10.1016/j.pss.2014.05.010.

Pajola, M., Lazzarin, M., Dalle Ore, C.M., Cruikshank, D.P., Roush,T.L., Magrin, S., Bertini, I., La Forgia, F., Barbieri, C., 2013a.Phobos as a D-type captured asteroid, spectral modeling from 0.25to 4.0 mm. Astrophys. J. 777, 1–6. https://doi.org/10.1088/0004-637X/777/2/127.

Pajola, M., Lazzarin, M., Dalle Ore, C.M., Cruikshank, D.P., Roush, T.L., Magrin, S., Bertini, I., La Forgia, F., Barbieri, C., 2013b. Phobosas a D-type captured asteroid, spectral modeling from 0.25 to 4.0 lm.Astrophys. J. 777 (127), 1–6. https://doi.org/10.1088/0004-637X/777/2/127.

Pavone, M., Castillo-Rogez, J.C., Nesnas, I.A.D., Hoffman, J.A., Strange,N.J., 2013. Spacecraft/Rover hybrids for the exploration of small solarsystem bodies. In: IEEE Aerospace Conf. Proc.. https://doi.org/10.1109/AERO.2013.6497160.

2186 A.N. Deutsch et al. / Advances in Space Research 62 (2018) 2174–2186

Peale, S.J., Canup, R.M., 2015. The origin of natural satellites. In: Treatiseon Geophysics, second ed., pp. 559–604. https://doi.org/10.1016/B978-0-444-53802-4.00177-9.

Pieters, C.M., Murchie, S., Thomas, N., Britt, D., 2014. Composition ofsurface materials on the moons of Mars. Planet. Space Sci. 102, 144–151. https://doi.org/10.1016/j.pss.2014.02.008.

Pieters, C.M., Taylor, L.A., Noble, S.K., Keller, L.P., Hapke, B., Morris,R.V., Allen, C.C., McKay, D.S., Wentworth, S., 2000. Space weath-ering on airless bodies: resolving a mystery with lunar samples.Meteor. Planet. Sci. 35, 1101–1107. https://doi.org/10.1111/j.1945-5100.2000.tb01496.x.

Pollack, J.B., Burns, J.A., 1977. An origin by capture for the martiansatellites? Bull. Am. Astron. Soc. 9, 518–519.

Pollack, J.B., Burns, J.A., Tauber, M.E., 1979. Gas drag in primordialcircumplanetary envelopes: a mechanism for satellite capture. Icarus37, 587–611. https://doi.org/10.1016/0019-1035(79)90016-2.

Ramsley, K.R., Head, J.W., 2013a. Mars impact ejecta in the regolith ofPhobos: bulk concentration and distribution. Planet. Space Sci. 87,115–129. https://doi.org/10.1016/j.pss.2013.09.005.

Ramsley, K.R., Head, J.W., 2013b. The origin of Phobos grooves fromejecta launched from impact craters on Mars: Tests of the hypothesis.Planet. Space Sci. 75, 69–95. https://doi.org/10.1016/j.pss.2012.10.007.

Ramsley, K.R., Head, J.W., 2014. Mars ejecta in the regolith of Phobos:implications for groove formation from secondary impacts. LunarPlanet. Sci. 45, abstract 1406.

Ramsley, K.R., Head, J.W., 2015. The secondary impact spike of Phobosfrom Stickney crater ejecta. Lunar Planet. Sci. 46, abstract 1201.

Ramsley, K.R., Head, J.W., 2017. The Stickney Crater ejecta secondaryimpact crater spike on Phobos: implications for the age of Stickneyand the surface of Phobos. Planet. Space Sci. 138, 7–24. https://doi.org/10.1016/j.pss.2017.02.004.

Rosenblatt, P., 2011. The origin of the martian moons revisited. Astron.Astrophys. Rev. 19, 19–44. https://doi.org/10.1007/s00159-011-0044-6.

Rosenblatt, P., Charnoz, C., 2012. On the formation of the martian moonsfrom a circum-martian accretion disk. Icarus 221, 806–815. https://doi.org/10.1016/j.icarus.2012.09.009.

Rosenblatt, P., Charnoz, S., Dunseath, K.M., Terao-Dunseath, M., Trinh,A., Hyodo, R., Genda, H., Toupin, S., 2016. Accretion of Phobos andDeimos in an extended debris disc stirred by transient moons. Nat.Geosci. 9, 581–583. https://doi.org/10.1038/ngeo2742.

Safronov, V.S., Pechernikova, G.V., Ruskol, E.L., Vitiazev, A.V., 1986.Protosatellite swarms. In: IAU Colloq. 77: Some Background aboutSatellites, pp. 89–116.

Sarafian, A.R., Nielsen, S.G., Marschall, H.R., McCubbin, F.M.,Monteleone, B.D., 2014. Early accretion of water in the inner solarsystem from a carbonaceous chondrite-like source. Science 346, 623–626. https://doi.org/10.1126/science.1256717.

Sasaki, S., 1990. Long-term orbital evolution of a captured satellite(Phobos) in a primary atmosphere. Meteoritics 25, 404.

Schmedemann, N., Michael, G.G., Ivanov, B.A., Murray, J.B., Neukum,G., 2014. The age of Phobos and its largest crater, Stickney. Planet.Space Sci. 102, 152–163. https://doi.org/10.1016/j.pss.2014.04.009.

Shi, X., Oberst, J., Willner, K., 2016. Mass wasting on Phobos triggeredby an evolving tidal environment. Geophys. Res. Lett. 43. https://doi.org/10.1002/2016GL071650, 2016GL071650.

Singer, S.F., 2007. Origin of the martian satellites Phobos and Deimos.First International Conference on the Exploration of Phobos andDeimos, abstract 7020.

Soter, S., Harris, A., 1977. Are striations on Phobos evidence for tidalstress? Nature 268, 421–422. https://doi.org/10.1038/268421a0.

Soter, S.L., 1971. The dust belts of Mars. Report 462, Center forRadiophysics and Space Research, Cornell Univ., Ithaca, New York.

Stooke, P.J., 2014. Mars sample return via robotic collection, Phoboscache and human retrieval. Lunar Planet. Sci. 45, abstract 1043.

Takir, D., Emery, J.P., McSween, H.Y., Hibbitts, C.A., Clark, R.N.,Pearson, N., Wang, A., 2013. Nature and degree of aqueous alterationin CM and CI carbonaceous chondrites. Meteorit. Planet. Sci. 48,1618–1637. https://doi.org/10.111/maps.12171.

Thomas, N., Stelter, R., Ivanov, A., Bridges, N.T., Herkenhoff, K.E.,McEwen, A.S., 2011. Spectral heterogeneity on Phobos and Deimos:HiRISE observations and comparisons to Mars Pathfinder results.Planet. Space Sci. 59, 1281–1292. https://doi.org/10.1016/j.pss.2010.04.018.

Thomas, P., Adinolfi, D., Helfenstein, P., Simonelli, D., Veverka, J., 1996.The surface of Deimos: contribution of materials and processes to itsunique appearance. Icarus 123, 536–556. https://doi.org/10.1006/icar.1996.0177.

Thomas, P.C., Veverka, J., Bloom, A., Duxbury, T.C., 1979. Grooves onPhobos: their distribution, morphology and possible origin. J. Geo-phys. Res. 84, 8457–8477. https://doi.org/10.1029/JB084iB14p08457.

Thomas, P.C., Veverka, J., Sullivan, R., Simonelli, D.P., Malin, M.C.,Caplinger, M., Hartmann, W.K., James, P.B., 2000. Phobos – Regolithand ejecta blocks investigated with Mars Orbiter Camera images. J.Geophys. Res. 105, 15091–15096. https://doi.org/10.1029/1999JE001204.

Truong, N., Lee, P., 2017. Origin of Phobos and Deimos by giant impact:lessons from terrestrial tektites. Lunar Planet. Sci. 48, abstract 3039.

Tsuda, Y., Yoshikawa, M., Abe, M., Minamino, H., Nakazawa, S., 2013.System design of the Hayabusa 2 – asteroid sample return mission to1999 JU3. Acta Astronaut. 91, 356–362. https://doi.org/10.1016/j.actaastro.2013.06.028.

Veverka, J., Duxbury, T.C., 1977. Viking observations of Phobos andDeimos: Preliminary results. J. Geophys. Res. 82, 4213–4223. https://doi.org/10.1029/JS082i028p04213.

Wanke, H., Dreibus, G., 1988. The Solar System: chemistry as a key to itsorigin – chemical composition and accretion history of terrestrialplanets. Phil. Trans. R. Soc. Lond. A 325, 545–557. https://doi.org/10.1098/rsta.1988.0067.

Weidenschilling, S.J., 1979. A possible origin for the grooves of Phobos.Nature 282, 697–698. https://doi.org/10.1038/282697a0.

Wilson, L., Head, J.W., 2015. Groove formation on Phobos: testing theStickney ejecta emplacement model for a subset of the groovepopulation. Planet. Space Sci. 105, 26–42. https://doi.org/10.1038/273282a0.